We’re launching Lyria 3.5 in Google Flow Music, with advances across musicality, lyrics, vocals, and creative control
deepmind.google·6d ago
AI Engineering
DeepMind Blog·Oct 23·5 min read
Gemini Robotics 1.5 brings AI agents into the physical world
TL;DR
Robots have struggled to effectively understand and interact with their environments for complex tasks. Gemini Robotics 1.5 introduces advanced AI models that enhance robots' ability to perceive, plan, and act.
✦ Why It Matters
Engineers can leverage Gemini Robotics 1.5 to develop more intelligent and capable robotic systems for complex tasks.
Key Takeaways
- Explore the Gemini API for advanced robotics development.
- Utilize Gemini Robotics-ER 1.5 for improved task planning.
- Implement vision-language-action capabilities in your robots.
- Leverage cross-embodiment learning to enhance robot adaptability.
How It Works
Gemini Robotics 1.5 combines two models: the VLA model translates visual and linguistic inputs into actions, while the embodied reasoning model plans and reasons about tasks. This collaboration allows robots to break down complex tasks into manageable steps, enhancing their ability to learn and adapt across different physical forms.